Keywords: speech production; model; fMRI; Broca’s area; premotor cortex; motor cortex; speech acquisition; sensorimotor learning; neural transmission delays This paper describes a neural model of speech acquisition and production that accounts for a wide range of acoustic, kinematic, and neuroimaging data concerning the control of speech movements. The model is a neural network whose components correspond to regions of the cerebral cortex and cerebellum, including premotor, motor, auditory, and somatosensory cortical areas. Computer simulations of the model verify its ability to account for compensation to lip and jaw perturbations during speech. Specific anatomical locations of the model’s components are estimated, and these estimates are used to simulate fMRI experiments of simple syllable production. 1 1

We present results from experiments on the temporal properties of prosodic events, providing evidence that accumulations of exemplars implicitly define perceptual target regions in prosody production. We argue that z-scores of segment and syllable durations are the relevant perceptual dimension of these regions. To support this hypothesis, we present experimental results confirming that realizations of segments and syllables in different prosodic contexts show significantly different z-score distributions. Further experiments show that the relationship between syllable z-scores and the z-scores of the corresponding segments is significantly stronger for infrequent than for frequent syllables. We claim that this is due to the fact that infrequent syllables have to be assembled from smaller units because they are not represented by enough exemplars to establish a syllable-level target region. 1.

The American English phoneme /r/ has long been associated with large amounts of articulatory variability during production. This paper investigates the hypothesis that the articulatory variations used by a speaker to produce /r/ in different contexts exhibit systematic tradeoffs, or articulatory trading relations, that act to maintain a relatively stable acoustic signal despite the large variations in vocal tract shape. Acoustic and articulatory recordings were collected from seven speakers producing /r/ in five phonetic contexts. For every speaker, the different articulator configurations used to produce /r/ in the different phonetic contexts showed systematic tradeoffs, as evidenced by significant correlations between the positions of transducers mounted on the tongue. Analysis of acoustic and articulatory variabilities revealed that these tradeoffs act to reduce acoustic variability, thus allowing relatively large contextual variations in vocal tract shape for /r/ without seriously ...

. This paper describes a neural model of speech production and perception-production interactions. This model has been developed to account for a wide variety of experimental data, ranging from kinematic analyses of articulator movements to functional imaging studies of the human brain. We have also tested predictions based on the model with these and other experimental techniques. Hypothesized neural correlates of the models components have been identified to facilitate testing of model predictions with techniques such as fMRI. The model also serves as a framework for interpreting and organizing the accumulating mass of data from functional imaging studies of the human brain. 1.

Abstract. This paper describes a neural model of speech production and perception-production interactions. This model has been developed to account for a wide variety of experimental data, ranging from kinematic analyses of articulator movements to functional imaging studies of the human brain. Hypothesized neural correlates of the model’s components have been identified to facilitate testing of model predictions with techniques such as fMRI. The model also serves as a framework for interpreting and organizing the accumulating mass of data from functional imaging studies of the human brain. According to the model, the goals of speech movements are in auditory-temporal space and the movements are planned with the use of mappings between articulations and their acoustic and auditory consequences. It is hypothesized that the mappings are acquired and maintained with the use of auditory feedback. Data are presented from studies of changes in speech that occur in response to a change in hearing status. These data provide information about the nature of the mappings and how they are used in planning speech movements. 1. The DIVA Model of Speech Production The overall objective of our research is to model the brain activity and the motor, biomechanical and sensory processes involved in speech production. Our approach is to use a combination of computational models and to develop and test them with brain imaging, psychophysical, physiological, anatomical and acoustic data. In particular, we have developed a neural network model of speech motor skill acquisition and speech production, called the DIVA model, that explains a wide range of data on contextual variability, motor equivalence,

A new paradigm for prosody research is presented, inspired by the speech production model recently proposed by Guenther, Perkell, and colleagues. This research paradigm aims at generalizing the production model by extending it from a predominantly segmental perspective to a new theory of the production of prosody. Speech movements in the prosodic domain are interpreted as intonational gestures that are planned to reach and traverse perceptual target regions. Evidence from F0 alignment studies suggests that the perceptual targets can be approximately represented by regions in a multidimensional acoustictemporal space. These studies also indicate that segmental, spectral, temporal, and prosodic structure are co-produced in such a way as to mutually support and enhance, and not impair, the perceptual targets. Furthermore, examples of multilevel mappings between invariant and variable targets in the domain of prosody are provided, and a dichotomy of phonemic and postural prosodic settings is discussed.

We investigate the structure of internal prosodic models. It is hypothesized that accumulations of perceived exemplars implicitly de ne prosodic target regions that are used in speech production. Results from two experimental studies on tonal and temporal events, respectively, are reported. The rst study concerns the categorical status of high and low boundary tones in German in a categorical perception paradigm. The aim of this experiment is to establish a methodology for testing the categorical status of prosodic events. In the second experiment we use z-scores as a measure of local speech rate. We demonstrate that this is an appropriate measure for the temporal dimension of perceptual target regions that are related to prosodic categories.

structural equation modeling, effective connectivity The neural substrates underlying auditory feedback control of speech were investigated using a combination of functional magnetic resonance imaging (fMRI) and computational modeling. Neural responses were measured while subjects spoke monosyllabic words under two conditions: (i) normal auditory feedback of their speech, and (ii) auditory feedback in which the first formant frequency of their speech was unexpectedly shifted in real time. Acoustic measurements showed compensation to the shift within approximately 135 ms of onset. Neuroimaging revealed increased activity in bilateral superior temporal cortex during shifted feedback, indicative of neurons coding mismatches between expected and actual auditory signals, as well as right prefrontal and Rolandic cortical activity. Structural equation modeling revealed increased influence of bilateral auditory cortical areas on right frontal areas during shifted speech, indicating that projections from auditory error cells in posterior superior temporal cortex to motor correction cells in right frontal cortex mediate auditory feedback control of speech. 1

Phonemic settings and the internal models that they represent are learned in the process of language and speech acquisition. Postural settings, in contrast, rely on continuous auditory monitoring and tend to break down quickly if this monitoring process is inhibited during speech production. Evidence presented in the literature seems to indicate that stable internal models are mostly associated with segmental phonemic targets, whereas prosodic features often display postural characteristics. In this paper it is argued that the dichotomy of phonemic and postural settings applies not only to segmental properties of speech but to prosodic features as well. Phonemic and postural effects on the production of prosody are reviewed and it is suggested that the boundary between phonemic and postural effects on a given prosodic feature is flexible. We further hypothesize that the speaker may rely on a set of acquired internal models and select from this set a particular model depending on communicative and situative constraints.

A model of the sensorimotor control of speech production is presented. The model is being implemented as a set of computer simulations. It converts an input sequence of discrete phonemes into quasi-continuous motor commands and a sound output. A key feature of the model is that the goals for speech movements, at least for some kinds of sounds, are regions in auditory-temporal space. The model is designed to have properties that are as faithful as possible to data from speakers – including measures of brain function, speech motor control mechanisms, physiology, anatomy, biomechanics and acoustics. Examples of simulations and actual data from some of these domains are presented. The examples demonstrate properties of the model or they are consistent with hypotheses generated from it. Our long-range goal is to implement the model completely and test it exhaustively, in the belief that doing so will significantly advance our understanding of speech motor control. 1.